Decision support for prediction and management of Long Covid Syndrome (LCS)

We will develop tools and knowledge to support physicians in accurately managing Long COVID syndrome (LCS) which has a significant impact on sufferers as well as their surroundings. Although much is now known regarding appropriate clinical management of acute COVID-19, very little is known about clinical manifestations, risk factors and underlying mechanisms for development of the highly heterogenous LCS. In this project, we aim to understand and mechanisms of LCS by combining front-line expertise from the fields of clinical medicine, virology, metabolism and immunology. We will study the pathogenesis of LCS by conducting geographically diverse cohort and registry studies, by conducting mechanistic studies, by using novel high-throughput methods for biomarker analysis, and by conducting interventional and follow-up studies on LCS patients. We will combine results from clinical and mechanistic studies to identify molecular and physiological parameters and/or pathways to decipher the mechanisms underlying LCS. We will exploit the high-throughput omics technologies to identify the predisposing factors and biomarkers that lead to the development of LCS. We will collect data from the cohort, mechanistic, biomarker and interventional studies and use these to validate the predictive artificial intelligence algorithms and to produce information and gain understanding on the combination of factors that lead to certain clustering of patients into different groups with specific symptoms. A machine learning and AI-informed Long Covid Prediction Support (LCPS) tool will be developed for the use of clinicians to predict the LCS and its possible clinical manifestations in patients. It will also help in the choice of personalized treatments for LCS patients. Additionally, an interactive graphic user interface infographic will also be available to clinicians and patients; this will communicate novel and understandable information about LCS and recommendations for patient management.